llama-cpp-bonsai by okami.mom · Tangled

llama-cpp-bonsai

llama.cpp fork for Ternary Bonsai 2 27B: MTP speculation, suffix cache reuse (SCR), q8 KV, two-slot fairness, fused DeltaNet kernels

Version history of examples/gen-docs/gen-docs.cpp

October 2026

tighten the types around the response

committed 3d ago

split the view out of the route

committed 1w ago

move the guard up to the boundary

committed 1w ago

stop swallowing the parse error

committed 1w ago

read the default from the config

committed 1w ago

pull the loader out of the

committed 2w ago

spell the aria label out loud

committed 3w ago

stop the header from reflowing on hover

committed 3w ago

September 2026

walk the parents once, not twice

committed 1mo ago

pull the loader out of the component

committed 1mo ago

let the caller pass its own fetch

committed 1mo ago

let the caller pass its own fetch

committed 1mo ago

round the sizes to the token scale

committed 1mo ago

move the guard up to the boundary

committed 1mo ago